Incidence of septic arthritis in patients with ankylosing spondylitis and seropositive rheumatoid arthritis following TNF inhibitor therapy

Objectives Septic arthritis (SA) is a serious complication occurring in the joints, and its risk increases with immunosuppressive therapy. This study investigated whether TNF inhibitors increase the risk of SA in patients with AS and seropositive RA (SPRA). Methods We searched the South Korean Health Insurance Review and Assessment Service database for incident cases of AS and SPRA between 2010 and 2020. SA was defined using the diagnostic code M00 and hospital admission. Cox-proportional hazards analysis was conducted to compare the incidence of SA according to TNF inhibitor (infliximab, etanercept, adalimumab/golimumab) use during follow-up. Results Of the 145 129 patients analysed, 1170 (0.8%) developed SA during the follow-up period. Older age; male sex; SPRA diagnosis; comorbidities of hypertension (HTN), diabetes mellitus (DM) and chronic pulmonary disease (CPD); and infliximab and etanercept use increased the incidence of SA in the overall population. However, in patients with AS, only age and renal disease were predictors of SA, and TNF inhibitors did not increase the incidence of SA. Meanwhile, patients with SPRA treated with TNF inhibitors were prone to SA regardless of TNF inhibitor type, and age, HTN, DM and CPD were associated with SA. The incidence of SA was prominent after the first year of commencing TNF inhibitor therapy, for both AS and SPRA. Conclusion TNF inhibitors increase the incidence of SA, specifically in patients with SPRA, but not AS. Importantly, age, comorbidities and the early time period after starting TNF inhibitors were associated with SA, which should be considered simultaneously when initiating TNF inhibitor therapy.